ML19098B054

From kanterella
Jump to navigation Jump to search

Non-Prop Enclosure 1: Final Safety Evaluation for Westinghouse Electric Company Topical Report WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0, Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio
ML19098B054
Person / Time
Site: Westinghouse
Issue date: 05/06/2019
From: Ekaterina Lenning
NRC/NRR/DLP/PLPB
To:
Lenning E, NRR/DLP, 415-3151
Shared Package
ML19094A033 -Pkg. List:
References
EPID L-2016-TOP-0015
Download: ML19098B054 (26)
v • d • e


Text

U. S. NUCLEAR REGULATORY COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION FINAL SAFETY EVALUATION FOR TOPICAL REPORT WCAP-18032-P, REVISION 0, AND WCAP-18032-NP, REVISION 0, "CALCULATION OF MIXED CORE SAFETY LIMIT MINIMUM CRITICAL POWER RATIO" WESTINGHOUSE ELECTRIC COMPANY Enclosure 1

TABLE OF CONTENTS 1.0 Introduction .................................................................................................................. 2.0 Regulatory Evaluation .................................................................................................. 3.0 Technical Evaluation .................................................................................................... 3.1 Review Framework for Critical Boiling Transition Models ............................................ Experimental Data ........................................................................................................ Model Generation ......................................................................................................... Model Validation ........................................................................................................... Validation Error..................................................................................................... Data Distribution ................................................................................................... 3.1.3.2.1 Validation Data ..................................................................................................... 3.1.3.2.2 Application Domain .............................................................................................. 3.1.3.2.3 Expected Domain ................................................................................................. 3.1.3.2.4 Data Density ......................................................................................................... 3.1.3.2.5 Sparse Regions .................................................................................................... 3.1.3.2.6 Restricted Domain ................................................................................................ Inconsistencies in the Models Error..................................................................... 3.1.3.3.1 Poolability ............................................................................................................. 3.1.3.3.2 Non-Conservative Subregions ............................................................................. 3.1.3.3.3 Model Trends ....................................................................................................... Quantified Model Error ......................................................................................... 3.1.3.4.1 Error Database ..................................................................................................... 3.1.3.4.2 Statistical Method ................................................................................................. 3.1.3.4.3 Appropriate Bias for Model Uncertainty ................................................................ Model Implementation .......................................................................................... 3.1.3.5.1 Same Computer Code .......................................................................................... 3.1.3.5.2 Same Evaluation Framework ............................................................................... 3.1.3.5.3 Transient Prediction ............................................................................................. 4.0 Limitations and Conditions ......................................................................................... 5.0 Conclusion ................................................................................................................. 6.0 Review of Sample License Amendment Request ...................................................... 7.0 List of Acronyms .........................................................................................................

1.0 INTRODUCTION

By letter dated December 12, 2016 (Ref. 1), Westinghouse Electric Company (WEC or Westinghouse) submitted topical report (TR) WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0 (also known as CENPD-300-P-A/CENPD-300-NP-A, Supplement 2), Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio to the U.S. Nuclear Regulatory Commission (NRC) for review and approval. The purpose of this report was to describe the methodology for calculating the safety limit minimum critical power ratio (SLMCPR) for boiling water reactors (BWRs) with mixed cores, that is, BWRs which contain Westinghouse and non-Westinghouse fuel (referred to as legacy fuel). This methodology is an update to Westinghouses currently approved methodology given in CENPD-300-P-A (Ref. 2).

The complete list of correspondence between the NRC and WEC is provided in Table 1 below.

This includes request for additional information (RAI) questions, responses to RAI questions, and any other correspondence relevant to this review.

Table 1: List of Key Correspondence Owner Document Document Date Reference WEC Submittal Letter December 12, 2016 1 Error!

Reference WEC Topical Report December 2016 source not found.

NRC RAI - Round 1 January 16, 2018 4 WEC RAI Responses May 11, 2018 5 Additionally, the NRC staff had five RAI questions. General information for each RAI questions including its number, its topic, its associated Goal, and the reference of its response are given in Table 2 below.

Table 2: Listing of RAIs Associated Reference of Reference of RAI Subject Goal RAI Response RAI-SNPB-01 Clarification of terminology Multiple 4 5 RAI-SNPB-02 Validation Error G3.4.2 4 5 RAI-SNPB-03 VIPRE Modeling G2 4 5 RAI-SNPB-04 Clarification of TR Multiple 4 5 RAI-SNPB-05 Example LAR G3.5 4 5 The methodology described in the TR is focused on generating critical power ratio (CPR) models for legacy fuel. Critical power is considered a special case of critical boiling transition (CBT).

2.0 REGULATORY EVALUATION

There are four main regulations associated with CBT:

  • Title 10 of the Code of Federal Regulations (10 CFR), Part 50, Appendix A, Criterion 10

Therefore, fuel failure is precluded during normal operation and AOOs.1 NUREG-0800, Standard Review Plan (SRP) Section 4.4, Thermal and Hydraulic Design includes the following two SAFDLs for use in accounting for the uncertainties involved in developing and using a CBT model (e.g., uncertainties in the values of process parameters, core design parameters, calculation methods, instrumentation) and ensuring that fuel failure is precluded:

(a) There should be a 95 percent probability at the 95 percent confidence level that the hot fuel rod in the core does not experience a CBT during normal operation or AOOs.

(b) At least 99.9 percent of the fuel rods in the core will not experience a CBT during normal operation or AOOs.

Typically, SAFDL (a) is associated with PWRs, and SAFDL (b) is associated with BWRs.

Before May 21, 1971, when the general design criteria (GDC) took effect, the Atomic Energy Commission (AEC), the predecessor to the NRC, approved construction permits for nuclear power plants based on AEC principal design criteria (PDC) that applicants proposed in their construction permit applications as required by the then-extant provisions of 10 CFR 50.34(a).

The AEC published proposed GDC in the Federal Register (32 FR 10213) on July 11, 1967, sometimes referred to as the AEC Draft GDC, which were generally consistent with the PDC previously proposed in applications for construction permits. AEC Draft GDC 6 is the relevant draft GDC which is substantially similar to the current GDC 10, and also calls for the reactor core to be designed with appropriate margin to specified limits which preclude fuel damage.

The second regulation associated with CBT is 10 CFR 50.36, part of which focuses on defining technical specification safety limits. There are multiple limits that are associated with CBT models used during plant operation. These limits can be operating limits, alarms, analysis limits, and safety limits. Generally, only the safety limit and associated limiting conditions for operation and surveillance requirements are included in the plants technical specifications. The safety limit associated with CBT is typically focused on an accurate quantification of the 1

Experiencing such a transition may not immediately result in fuel failure. The decrease in heat transfer and subsequent increase in fuel temperature may not be enough to cause the cladding to weaken or melt. Therefore, the point of CBT is considered to be a conservative limit compared to the actual point of fuel damage.

uncertainty of the CBT model and may also include the quantification of additional uncertainties as well.

The third regulation associated with a CBT is in 10 CFR 50.34, which focuses on defining the information that a licensee must present to ensure safe operation. Specifically, 10 CFR 50.34(a)(4) requires that the Preliminary Safety Analysis Report include determination of the margins of safety during normal operation and AOOs. One of these is the margin to CBT, which verifies that fuel failure is precluded during normal operation and AOOs through analysis.

The fourth regulation associated with a CBT appears in 10 CFR Part 50, Appendix B. It requires licensees to include certain structures, systems, and components (SSCs) in a quality assurance program that satisfies specific criteria. Appendix B, Criterion III, requires that specified design control measure be applied to the design of safety-related SSCs, and these measures apply to safety analyses for these SSCs. The CBT model is a key component of the safety analysis subject to 10 CFR Part 50, Appendix B.

The NRC staff has performed its review according to detailed criteria for CBT models that derive from experience accumulated over a number of previous reviews in this topical area. Further details about those criteria can be found in NUREG/KM-0013 (Ref. 15).

3.0 TECHNICAL EVALUATION

WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0, describes how the SLMCPR will be calculated for a core containing both Westinghouse fuel and non-Westinghouse fuel (i.e., legacy fuel). Such mixed cores are common when plants transition from one fuel vendor to another. The methodology described by Westinghouse does not modify how the SLMCPR will be calculated for a core loaded with 100 percent Westinghouse fuel. Rather, it provides an approach to generate a CPR model for the legacy fuel, so Westinghouse can use that model in the SLMCPR evaluation for the mixed core.

The NRC staffs technical evaluation is focused on determining if the CPR model generated using this approach is acceptable for use in reactor safety license calculations (i.e., that the model can be trusted). To perform this evaluation, the NRC staff used a framework similar to the framework used in the NRC staffs safety SE of the ORFEO-CHF correlation (Ref. 6), the ACE/ATRIUM-11 CPR correlation (Ref. 7), and the NuScale Power CHF correlation (Ref. 8).

However, application of this framework review differs from the previous reviews in that instead of evaluating a single CBT model the NRC staff is evaluating a methodology to generate a specific type of CBT model, a CPR model for legacy fuel. Thus, the CBT models generated using this methodology will have limited applicability and as they will not be used to predict CPR on fresh fuel assemblies. Generally, 2nd and 3rd cycle assemblies are less limiting than fresh assemblies and accordingly contribute less to the SLMCPR value. Hence, the criteria for the CPR models used to predict legacy fuel are generally less strict than those for models which are used to predict fresh fuel, and therefore the NRC staff believes it is reasonable to approve a methodology for addressing legacy fuel instead of requiring legacy fuel models to be submitted for each use.

3.1 Review Framework for Critical Boiling Transition Models This section discusses the review framework for CBT models that was applied in this review.

This framework is expressed using concepts from goal structuring notation (GSN). In GSN, the

safety case is presented by as a structure which contains multiple goals. The top goal, is a high-level statement we wish to demonstrate as true. It would be very difficult to demonstrate this statement is true with some set of basic evidence. Therefore, the top goal is decomposed into a set of goals (i.e., sub-goals). In this decomposition, proving each sub-goal is true is considered equivalent to proving the top goal is true. Further, each sub-goal is further decomposed, and so on, until a set of goals are obtained which can be demonstrated to be true through some basic evidence. For clarity, this last set of goals which are demonstrated to be true via evidence are termed base goals.

For CBT models, the top goal is: The critical boiling transition model can be trusted in reactor safety analyses. Based on the NRC staffs experience reviewing these models, a study of previous SEs, and multiple discussions with various industry experts, this goal is decomposed into the three sub-goals given in Figure 1.

Figure 1: Decomposing G - Main Goal These sub-goals are discussed in the sections below.

Experimental Data Experimental data are the cornerstone of a CBT model. Demonstrating the experimental data are are appropriate is generally accomplished using the three sub-goals given in Figure 2 below.

Figure 2: Decomposing G1 - Experimental Data For the methodology described in WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0, the decomposition of G1 into the sub-goals is not necessary. The data used to support the model is either data from the legacy fuel vendors test facility or from the legacy fuel vendors CPR model. In either instance, that data and/or that model have been previously approved by the NRC staff. Because Westinghouse will be using data which has been used to generate a previously approved CPR model or will be using predictions from an approved CPR model, the NRC staff has determined that the data supporting the model is appropriate and goal G1 has been met.

Model Generation There are numerous ways to generate a CBT model. Demonstrating the model generation is appropriate is accomplished using the two sub-goals given in Figure 3 below.

Figure 3: Decomposing G2 - Model Generation

For the methodology described in WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0, the decomposition of G2 into the sub-goals is not necessary. As specified in the TR, the model form will be based on an approved CPR model, which will be denoted in this evaluation by the identifier X (representing an alphanumeric identifier for the CPR model, e.g., the D4 correlation). CPR models which are generated to predict the performance of legacy fuels and are developed based on an approved CPR model will be denoted by the base model identifier modified with prime marks (e.g., X, X, X). The method used to develop the legacy fuel CPR model depends on whether the data provided to Westinghouse by the legacy fuel vendor represents (1) a limited CPR database, (2) an extensive (but incomplete) CPR data base, or (3) a complete CPR data base. In Cases 1 and 2, it is expected that CPR data provided will be generated by the NRC-approved CPR correlation for the legacy fuel. In Case 3, it is expected that a full database will contain the data from the CPR experiments used to develop the legacy fuels NRC-approved CPR correlation.

Case 1 - Limited CPR Database

[

]

[

]

Case 2 - Extensive, But Incomplete CPR Database

[

]

[

]

Case 3 - Complete CPR Database

[

]

Summary Starting with an NRC-approved Westinghouse CPR model, X, Westinghouse will generate one or more of the following models depending one or more the legacy fuel CPR data available:

(1) Case 1 - X - [

]

(2) Case 2 - X - [

]

(3) Case 3 - X - [

]

Because Westinghouse will be starting with a previously approved CPR models mathematical form, applying a method to correlate R-factors and other model coefficients which is consistent with their currently approved approach, and potentially applying a method to optimizing model coefficients based on legacy data which is also consistent with their currently approved approach, the NRC staff has determined that the model has been generated in a logical fashion and goal G2 has been met.

Model Validation Validation is the accumulation of evidence which is used to assess the claim that a model can predict a real physical quantity (Ref. 9). Thus, validation is a never-ending process as more evidence can always be obtained to bolster this claim. However, at some point, when the accumulation of evidence is considered sufficient to make the judgment that the model can be

trusted for its given purpose, the model is said to be validated. Demonstrating the model validation is appropriate is accomplished using the four sub-goals given in Figure 4 below. Further details for each of these goals can be found in the NUREG/KM-0013 (Ref. 15).

Figure 4: Decomposing G3 - Model Validation In the submittal, Westinghouse has not provided the validation data for a specific legacy fuel CPR model but has proposed a process which will be used to validate such models in the future. In general, CPR models are submitted to the NRC for review and approval, and the validation of the model is typically the NRC staffs main focus during the review. In the past, however, the NRC staff has not used this process to review the application of NRC-approved critical power models in determining the SLMCPR safety limit for cores containing co-resident fuel from another fuel vendor2. This is because, typically, the co-resident fuel is in its second cycle of operation or later. As a result, this legacy fuel generally does not have the impact on the SLMCPR limit that fresh fuel has.

The validation of each legacy fuel CPR model will need to be confirmed by Westinghouse, and certain information should be provided to the NRC staff in the license amendment request (LAR) which would accompany the application of the model. The following section discusses the validation activities which should occur, and what information should be submitted in the LAR.

This section is meant to provide the technical background for validation. For a complete understanding of what information should be included in the LAR, see Section 6, Review of Sample License Amendment Request, of this SE.

2 The NRC staff has previously reviewed methodologies that call for the modification of one vendors CPR correlation to evaluate co-resident fuel from another vendor in order to determine the safety limit (e.g.,

Framatomes EMF-2245 (Ref.11)). Additionally, the NRC staff has previously approved a method for updating the coefficients of a CPR model without prior NRC review i.e., General Electrics NEDO-24011-A, Revision 18 (Ref.12)).

Validation Error Validation Error The correct validation error has been calculated.

G3.1, Review Framework for Critical Boiling Transition Models For a critical power correlation, the validation error should be based on the critical power from the experimental tests and the predicted critical power of the test assembly at the experiments conditions. However, for the legacy fuel, the actual experimental data will not be available with the exception of the rare instances of Case 3 (as defined above). Instead, the legacy fuel CPR data is used in place of the experimental data.

Westinghouse has identified that the data used to generate the validation error will be from the legacy fuel vendor. In general, this will be data generated using that vendors approved CPR model (Cases 1 and 2) or from the legacy fuel CPR experimental data (Case 3). Because that data is the best estimate for the true CPR value, the NRC staff has determined that this goal has been met.

Data Distribution The second sub-goal in demonstrating that the models validation is appropriate is to demonstrate demonstrate that the data is appropriately distributed throughout the application domain. This is typically demonstrated using the six sub-goals as given in Figure 5 below.

Figure 5: Decomposing G3.2 - Data Distribution

No further decompositions of the sub-goals were deemed useful. Therefore, the evidence demonstrating the following goals were met is provided below.

3.1.3.2.1 Validation Data Validation Data The validation data (i.e., the data used to quantify the models error) should be identified.

G3.2.1, Review Framework for Critical Boiling Transition Models In Appendix E of the TR, Westinghouse identified that for Cases 1 and 2, all of the legacy fuel data would be used as validation data, and that for Case 3, some number of data points would be reserved for validation purposes. Because Westinghouse has identified the validation data, the NRC staff has determined that this goal has been met.

3.1.3.2.2 Application Domain Application Domain The application domain of the model should be mathematically defined.

G3.2.2, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR. In section 3 of that sample submittal, Westinghouse provided a summary of the data used to validate the model (i.e., legacy state points).

Therefore, the NRC staff expects that this same type of information will be provided in a plant-specific licensing submittal which references this TR for the calculation of a mixed core SLMCPR.

In particular, Westinghouse may satisfy this goal by providing information in a licensing submittal consistent with the sample submittal provided in the RAI response (see SEs Section 6, Review of Sample License Amendment Request, Item 2) and further confirming that validation data covers the range of normal operation and AOOs (see SEs Section 6, Review of Sample License Amendment Request, Item 1.a).

3.1.3.2.3 Expected Domain Expected Domain The expected domain of the model should be understood.

G3.2.3, Review Framework for Critical Boiling Transition Models

The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR; however, information on this goal was missing from the sample submittal.

Generally, the expected domain for a CPR model is defined by the following figures.

- Pressure vs. Mass Flux

- Pressure vs. Subcooling

- Pressure vs. R-factor

- Subcooling vs. Mass Flux

- R-factor vs. Mass Flux

- Subcooling vs. R-factor Westinghouse may satisfy this goal by providing confirmation in the licensing submittal that the above plots were generated and appropriately analyzed by Westinghouse (see SEs Section 6, Review of Sample License Amendment Request, Item 1.b).

3.1.3.2.4 Data Density Data Density There should be adequate data density throughout the expected and application domains.

G3.2.4, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR, however, information on this goal was missing from the sample submittal.

The NRC staff expects that Westinghouse will ensure the expected and application domains have adequate data density. Westinghouse may satisfy this goal by providing confirmation in the licensing submittal that both domains were examined and do have adequate data density (see SEs Section 6, Review of Sample License Amendment Request, Item 1.c).

3.1.3.2.5 Sparse Regions Sparse Regions Sparse regions (i.e., regions of low data density) in the expected and application domains should be identified and justified to be appropriate.

G3.2.5, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In its response

to RAI-05, Westinghouse provided a sample submittal for such an application of the TR, however, information on this goal was missing from the sample submittal.

The NRC staff expects that Westinghouse will ensure there are no sparse regions in the expected and application domains, or if such regions exist, it is appropriate to use the CPR model in those regions. Westinghouse may satisfy this goal by providing confirmation in the licensing submittal that both domains were examined and either do not have sparse regions, or the use of the model in the sparse region is justified (see SEs Section 6, Review of Sample License Amendment Request, Item 1.d).

3.1.3.2.6 Restricted Domain Restricted Domain The model should be restricted to its application domain.

G3.2.6, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR; however, information on this goal was missing from the sample submittal.

The NRC staff expects that Westinghouse will ensure that the Westinghouse generated legacy CPR model will not be applied outside of the domain over which it was validated. Westinghouse may satisfy this goal by providing confirmation in the licensing submittal that the model will not be applied outside the domain over which it was validated (see SEs Section 6, Review of Sample License Amendment Request, Item 1.e).

Inconsistencies in the Models Error The third sub-goal in demonstrating that the models validation was appropriate is to demonstrate demonstrate that the model error is consistent over the application domain. This is typically demonstrated using the three sub-goals as given in Figure 6 below.

Figure 6: Decomposing G3.3 - Inconsistencies in the Models Error No further decompositions of the sub-goals were deemed useful. Therefore, the evidence demonstrating the following goals were met is provided below.

3.1.3.3.1 Poolability Poolability The validation error should be investigated to determine if it contains any sub-groups which are obviously not from the same population (i.e., not poolable).

G3.3.1, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05 (Ref. 5), Westinghouse provided a sample submittal for such an application of the TR; however, information on this goal was missing from the sample submittal.

The NRC staff expects that Westinghouse will ensure that the validation error did not contain any subgroups which were obviously not from the same population. Westinghouse may satisfy this goal by providing confirmation in the licensing submittal that there are no obviously non-poolable subgroups (see SEs Section 6, Review of Sample License Amendment Request, Item 1.f).

3.1.3.3.2 Non-Conservative Subregions Non-Conservative Subregions The expected domain should be investigated to determine if it contains any non-conservative subregions which would impact the predictive capability of the model.

G3.3.2, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR, however, information on this goal was missing from the sample submittal.

The NRC staff expects that Westinghouse will ensure that the expected domain does not contain any non-conservative subregions which would impact the predictive capability of the model. Westinghouse may satisfy this goal by (1) providing confirmation in the licensing submittal that there are no non-conservative subregions in the expected domain (see Section 6

- Review of Sample License Amendment Request, Items 1.g) and (2) including figures describing the error and demonstrating those figures demonstrate no adverse trends, such as the figures E1 - E6 of Appendix E (see SEs Section 6, Review of Sample License Amendment Request, Item 4).

3.1.3.3.3 Model Trends Model Trends The model is trending as expected in each of the various model parameters.

G3.3.3, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In their response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR. In section 3 of that sample submittal, Westinghouse provided trends of the models predictions in each of the important model parameters.

The NRC staff expects that Westinghouse will ensure that the model is trending as expected in each of the various model parameters. Westinghouse may satisfy this goal by providing figures similar to those in the sample submittal and confirming those figures demonstrate the expected behavior, as do the figures in the sample submittal (see SEs Section 6, Review of Sample License Amendment Request, Item 3).

Quantified Model Error The fourth sub-goal in demonstrating that the models validation was appropriate is to demonstrate that the model error has been appropriately quantified over the application domain.

This is typically demonstrated using the three sub-goals as given in Figure 7 below.

Figure 7: Decomposing G3.4 - Quantified of the Models Error No further decompositions of the sub-goals were deemed useful. Therefore, the evidence demonstrating the following goals were met is provided below.

3.1.3.4.1 Error Database Error Database The validation error statistics should be calculated from an appropriate database.

G3.4.1, Review Framework for Critical Boiling Transition Models In RAI-02, the NRC staff requested a further analysis of the error of the X and X models, and if their error when compared to the legacy fuel data would bound their error relative to the true CPR value. Westinghouse responded to RAI-02 (Ref. 5) by providing an analysis which

[

]

Westinghouse identified that the legacy fuel data will be used for validating the model. That data is either from the legacy fuel CPR model (Cases 1 and 2) or from the legacy fuel CPR experimental data (Case 3). Because that data is the best estimate for the true CPR value, the NRC staff has determined that this goal has been met.

3.1.3.4.2 Statistical Method Statistical Method The validation error statistics should be calculated using an appropriate method.

G3.4.2, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In its response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR; however, information on this goal was missing from the sample submittal.

In response to RAI-02, Westinghouse provided a detailed explanation of the error values that will be used for each of the three cases described above. [

]

The NRC staff expects that Westinghouse will identify which of the above approaches was used to calculate the validation error. Westinghouse may satisfy this goal by providing information in their licensing submittal specifying which of the approaches was used (see Section 6, Review of Sample License Amendment Request, Item 5).

3.1.3.4.3 Appropriate Bias for Model Uncertainty Appropriate Bias The models error should be appropriately biased in generating the model uncertainty.

G3.4.3, Review Framework for Critical Boiling Transition Models The NRC staff determined that this goal could only be satisfied by information that would be developed during an application of the TR to generate a legacy fuel CPR model. In its response to RAI-05, Westinghouse provided a sample submittal for such an application of the TR; however, information on this goal was missing from the sample submittal.

The NRC staff expects that Westinghouse will ensure that the model will be appropriately biased. Westinghouse may satisfy this goal by providing confirmation in the licensing submittal that the model needed no bias or providing adequate discussion of what bias was used and how such a bias was generated (see Section 6 - Review of Sample License Amendment Request, Item 1.h).

Model Implementation The fifth sub-goal in demonstrating that the models validation was appropriate is to demonstrate demonstrate that the model will be implemented in a manner consistent with its validation. This is is typically demonstrated using the three sub-goals as given in Figure 8 below:

Figure 8: Decomposing G3.5 - Model Implementation

No further decompositions of the sub-goals were deemed useful. Therefore, the evidence demonstrating the following goals were met is provided below.

3.1.3.5.1 Same Computer Code Same Computer Code The model has been implemented in the same computer code which was used to generate the validation error.

G3.5.1, Review Framework for Critical Boiling Transition Models Because critical power models do not rely on subchannel methods (which are code specific) and because implementation of the Westinghouse-generated legacy CPR model will be validated with the same code as that which was used to validate the Westinghouse generated legacy CPR model, the NRC staff has determined that this goal is not applicable.

3.1.3.5.2 Same Evaluation Framework Same Evaluation Framework The model used for prediction of the Critical Boiling Transition is applied in the same manner as it was when determining the models error from the validation error set.

G3.5.2, Review Framework for Critical Boiling Transition Models Unlike DNB models, which are calculated in a subchannel code with a number of complex closure relations, the application of a CPR model is relatively simple. A single channel is modeled with R-factors accounting for radial powers and additive constants accounting for thermal hydraulic effects. Given the simplicity of CPR models and Westinghouses previous description of the procedures for implementation and validating such a model in a new computer code (response to RAI-SNPB-34 from a previous review (Ref. 14)), the NRC staff has determined that this goal has been met.

3.1.3.5.3 Transient Prediction Transient Prediction The model results in an accurate or conservative prediction when it is used to predict transient behavior.

G3.5.3, Review Framework for Critical Boiling Transition Models Because CPR models are generated using steady state data, they are commonly evaluated against transient experimental data to confirm that a steady state model can accurately (or conservatively) predict critical power during a transient. For the legacy fuel CPR models

addressed in this TR, there is no transient data available for such a comparison. However, because both the underlying legacy fuel CPR model and the original Westinghouse model (X) will have been confirmed to accurately or conservatively predict transient data, the NRC staff determined that it is reasonable to assume that the Westinghouse-generated legacy fuel CPR model (X, X, or X) will also accurately or conservatively predict transient data. Additionally, the error used on the Westinghouse-generated legacy fuel CPR model will be demonstrated to bound the legacy fuel data as discussed in Section 3.1.3.4.2, Statistical Method, of this SE.

For those instances when sufficient legacy fuel data is not available to ensure the error is adequately quantified in the transient domain, [

]

Because the error of the Westinghouse-generated legacy fuel CPR model will bound the legacy fuel CPR value in the transient domain, or be demonstrated to conservatively predict the behavior, the NRC staff has determined that this goal has been met.

4.0 LIMITATIONS AND CONDITIONS

1. One of the following conditions must be met for the use of the Westinghouse legacy fuel CPR model:

[

]

[

]

5.0 CONCLUSION

Based on the review described in Section 3.1.1, Experimental Data, of this SE, the NRC staff concludes that the experimental data supporting the Westinghouse-generated legacy fuel CPR model will be appropriate. Based on the review described in Section 3.1.2, Model Generation, of this SE, the NRC staff concludes that the Westinghouse generated legacy fuel CPR model will be generated in a logical fashion. Based on the review described in Section 3.1.3, Model Validation, of this SE, the NRC staff concludes that the Westinghouse-generated legacy fuel CPR model will have sufficient validation as demonstrated through appropriate quantification of its error. If the methodology described by Westinghouse and discussed above is followed, which will be confirmed in a future NRC staff review, the NRC staff concludes that the Westinghouse-generated legacy fuel CPR model may be trusted in reactor safety analysis subject to the Limitations and Conditions listed in Section 4.0, Limitations and Conditions, of this SE.

6.0 REVIEW OF SAMPLE LICENSE AMENDMENT REQUEST In response to RAI-05, Westinghouse provided a sample license amendment request which would be submitted with the use of a Westinghouse-generated CPR model for legacy fuel. With the information provided in the example submittal, along with some additional information described below, such a submittal should contain the information needed by the NRC staff to perform a review of the implementation of a Westinghouse-generated CPR model for legacy fuel. Therefore, the NRC staff recommends that future licensing actions implementing WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0, Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio, address the following items:

1. In addition to the example submittal provided in the TR, in their actual LAR Westinghouse should confirm the following statements are accurate:
a. That Westinghouse has examined the data used to validate the model (e.g., legacy state points) and that the data covers the range of normal operation and AOOs.
b. That Westinghouse has generated the necessary plots to understand the expected domain.
c. That Westinghouse has examined the data density in the expected and application domains and determined that it was adequate.
d. That Westinghouse has examined the expected domain and application domain and determined that there were no sparse regions, or that any sparse regions could be justified.
e. That Westinghouse has restricted the use of the legacy fuel CPR model to the domain over which it was validated.
f. That Westinghouse has analyzed the validation error and the error did not contain any sub-groups which were obviously not from the same population.
g. That Westinghouse has analyzed the validation error and the error did not contain any non-conservative subregions.
h. That Westinghouse has either applied an appropriate bias or determined that no bias was required.
2. The application domain of the CPR model should be provided and be adequately covered by the validation data. In the example submittal in the TR, the information provided in the bulleted list in section F.3 should generally be sufficient.
3. The trends of the Westinghouse-generated legacy fuel CPR model should be similar to the example submittal in the TR, Figures F-3 through F-6.
4. Westinghouse should provide graphs which describe the error in the Westinghouse-generated legacy fuel CPR model. The following graphs should be provided.
a. Histogram of prediction error
b. Prediction error vs. pressure
c. Prediction error vs. mass flux
d. Prediction error vs. sub-cooling
e. Prediction error vs. I2
f. Prediction error vs. R-factor The histogram should look normally distributed. The prediction error vs. each parameter should not contain any non-conservative trends. For examples of reasonable plots, see Figures E-13 to E-18 and Figures E-19 to E-24 of appendix E in the TR.
5. The NRC staff should confirm that the error of the Westinghouse generated legacy fuel CPR model satisfies the conditions and limitations of this SE.

If the NRC staff reviewing the submittal can confirm the above are true, then the Westinghouse generated CPR model for legacy fuel should be considered acceptable for use in reactor safety analysis.

7.0 LIST OF ACRONYMS AEC Atomic Energy Commission AOO anticipated operational occurrence BWR boiling water reactor CBT critical boiling transition CFR Code of Federal Regulations COLR core operating limits report CPR critical power ratio GDC general design criteria NRC U. S. Nuclear Regulatory Commission PWR pressurized water reactor RAI request for additional information SAFDL specified acceptable fuel design limit SE safety evaluation SLMCPR safety limit minimum critical power ratio SRP standard review plan SSCs structures, systems, and components TR topical report

24 REFERENCES

1. Letter from Gresham, J., Westinghouse Electric Company (Westinghouse), to U.S.

Nuclear Regulatory Commission (NRC), Submittal of WCAP-18032-P, Revision 0 and WCAP-18032-NP, Revision 0, Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio (Proprietary/Non-Proprietary), LTR-16-79, December 12, 2016 (ADAMS Accession No. ML16350A113).

2. Westinghouse, Reference Safety Report for Boiling Water Reactor Reload Fuel, CENPD-300-P-A, July 1996 (ADAMS Accession No. ML003767366).
3. Letter from Hsueh, K., NRC, to J. A. Gresham, Westinghouse, Acceptance for Review of Westinghouse Electric Company Topical Report WCAP-18032-P, Revision 0 and WCAP-18032-NP, Revision 0, Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio, February 17, 2017 (ADAMS Accession No. ML17041A335).
4. Letter from Lenning, E., NRC, to Gresham, J., Westinghouse, Request for Additional Information Re: Westinghouse Electric Company WCAP-18032-P, Revision 0, and WCAP-18032-NP, Revision 0, Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio Topical Report (TAC NO. MF9017), January 16, 2018 (ADAMS Accession No. ML17352A013).
5. Letter from Gresham, J., Westinghouse, to NRC, Responses to NRC Request for Additional Information for Westinghouse Topical Report WCAP-18032-P/WCAP-18032-NP, Revision 0, Calculation of Mixed Core Safety Limit Minimum Critical Power Ratio (Proprietary/Non-Proprietary), LTR-18-28, May 11, 2018 (ADAMS Accession No. ML18135A063).
6. Safety Evaluation for Topical Report-ANP-10341P, The ORFEO-GAIA and ORFEO-NMGRID Critical Heat Flux Correlations, September 24, 2018 (ADAMS Accession No. ML18236A426).
7. Safety Evaluation for Topical Report-ANP-10335P, Revision 0, ACE/ATRIUM 11 Critical Power Correlation, May 25, 2018 (ADAMS Accession No. ML18207A408).
8. Safety Evaluation for Topical Report-0116-21012, Revision 1, NuScale Power Critical Heat Flux Correlations, August 8, 2018 (ADAMS Accession No. ML18214A480).
9. Kaizer, J.S, Identification of Non-Conservative Subregions in Empirical Models Demonstrated Using Critical Heat Flux Models, Nuclear Technology, Vol. 190, 65-71, 2015.
10. Oberkampf, W.L and Roy, C.J., Verification and Validation in Scientific Computing, Cambridge University Press, Cambridge (2010).
11. Siemens, Application of Siemens Power Corporation's Critical Power Correlations to Co-Resident Fuel, EMF-2245(NP)(A), Revision 0, August 2000 (ADAMS Accession No. ML003753200).

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

12. General Electric, General Electric Standard Application for Reactor Fuel (GESTAR II, Main), NEDO-24011-A, Revision 18, April 2011 (ADAMS Accession No. ML111120046).
13. Box, G.E.P, Hunter, W.G., and Hunter, J.S., Statistics for Experimenters, John Wiley &

Sons, New York (1978).

14. Letter from Gresham, J., Westinghouse, to NRC, Responses to NRC Request for Additional Information for Westinghouse Electric Company Topical Report WCAP-17794-P/WCAP-17794-NP, Revision 0, 10x10 SVEA Fuel Critical Power Experiments and New CPR Correlation: D5 for SVEA-96 Optima3 Fuel (Proprietary/Non-Proprietary), LTR-16-53, August 8, 2016 (ADAMS Accession No. ML16224B087).
15. Kaizer, J.S., Anzalone, R., Brown, E., Panicker, M., Haider, S., Gilmer, J., Drzewiecki, T., and A. Attard. Credibility Assessment Framework for Critical Boiling Transition Models, DRAFT for comment NUREG/KM-0013, 2019.

Principal Contributor: J. Kaizer, NRR/DSS Date: May 6, 2019

Retrieved from "https://kanterella.com/w/index.php?title=ML19098B054&oldid=1910872"